Maiga 2013

Electronic Government, An International Journal, Vol. 10, Nos.
3/4, 2013 259
Adoption of e-tax services in Uganda: a model of

citizen-based factors
Gilbert Maiga* and Elizabeth Asianzu

Department of Informatics Technology,
Makerere University,
Box 7062, Kampala, Uganda
E-mail: gmaiga@cit.ac.ug
E-mail: easianzu@yahoo.com
*Corresponding author
Abstract: Diverse benefits of e-government services are linked to its adoption

and usage. E-government adoption rates in economically transiting countries
remain low and so its benefits are not fully realised. This is partly due to
governments focusing largely on technical supply-side factors with little
emphasis on the demand or citizens’ perspective of e-government adoption.
The result has been a gap between what is offered and what is consumed. This
paper presents the results of a study that develops a model for e-tax service
adoption as an attempt to bridge the gap for this segment of e-government.
Requirements for the model elicited in a field study are used to extend the
technology acceptance model (TAM) which emphasises citizen-based factors
for e-tax adoption. The extended model has dimensions of adoption benefits,
trust, attitudes, education, compatibility, awareness, accessibility, training, user
support and local language use. It is generic and reusable for other countries in
similar context.
Keywords: e-government; e-tax adoption; e-government adoption models;

citizen-based factors; TAM; technology acceptance model; transitioning
country; developing country.
Reference to this paper should be made as follows: Maiga, G. and Asianzu, E.

(2013) ‘Adoption of e-tax services in Uganda: a model of citizen-based
factors’, Electronic Government, An International Journal, Vol. 10, Nos. 3/4,
pp.259–283.
Biographical notes: Gilbert Maiga has a PhD in Information Systems from

Makerere University. His PhD was in the area of Biomedical Ontology
Evaluation. He is a Senior Lecturer and Chair of the Department of Information
Technology at Makerere University. His main areas of interest for research are
medical information systems and e-services.
Elizabeth Asianzu holds a Bachelor of Administrative and Secretarial Science

(KYU) and Postgraduate Diploma in ICT Policy and Regulations. She is
currently pursuing a Master of Information Technology degree from Makerere
University.
Copyright © 2013 Inderscience Enterprises Ltd.

260 G. Maiga and E. Asianzu
1 Introduction
Whereas e-government adoption has increased for most countries, in economically

and technologically transitioning countries, it remains constrained by low adoption rates.
E-government usage and realisation of its benefits is therefore far from reaching
maximum potential until the gap between what is offered and what is used is bridged
(Al-Adawi et al., 2005). The unrealised potential of e-government is partly because
governments have focused more on the supply side and disregarded the demand or
citizens perspective (Arslan, 2009). The role of the demand-side factors in bridging the
gap between what is offered and what is used has not been emphasised, making citizens
in developing countries lag behind in adoption of e-services (Moon and Kim, 2001).
The concept of e-government revolves largely around the citizen. In order to develop
this citizen focused vision, policymakers must keep the ordinary citizen in mind when
designing systems because e-government is not just a cost cutting or efficiency initiative,
but rather is directed at bettering the lives of ordinary people (Infodev, 2002). The focus
for e government initiatives in the developing countries has, however, largely been
technological by bureaucrats who view the citizen as neither a customer of government
nor a participant in decision-making.
In Uganda, there is low utilisation of ICT during the delivery of various public
services, low utilisation of e-services as well as inadequate research and development in
IT to inform government’s uptake of relevant applications (MoICT, 2009). Furthermore,
despite efforts to develop public administration through the e-government projects,
no tangible-positive results have been realised to date. This is partly because
e-government adoption issues have mostly addressed technical issues (supply-side
factors) of infrastructure, legislation, policy and process. Government agencies do not as
a rule engage citizens in the development of their e-government services and resources.
Rather, many applications are internally driven to meet cost savings and other
government mandates regarding efficiency (Bertot et al., 2008). The role of non-technical
(citizen-based) factors (e.g., trust, attitudes, education, awareness, accessibility, training,
user support, use of local language) in the adoption of e-government has not been
emphasised leading to a low usage level of ICT and internet in Uganda (Senne, 2005).
To become efficient, governments require a citizen-centric model that involves key
stakeholders outside of government – businesses, trade associations, scientists, academics
and NGOs. Without their input, e-government projects are unlikely to succeed, as citizens
will not use a system that does not respond to their needs (Infodev, 2002). In Uganda,
while attention has been focused on the technical issues of e-tax services adoption like
infrastructure, legislation, policy and process, the role of non-technical (consumer-based)
factors in the adoption of e-tax services has not been emphasised. The result has been a
low usage level of ICT and e-tax services in Uganda (Senne, 2005) with existing installed
e-tax systems not being optimally utilised (Sserwanga, 2010). Thus, the question of what
citizen-based factors or model can best support bridging the gap between the availability
of e-tax services and its adoption and usage in Uganda remains largely unanswered. The
literature therefore points to unresolved questions for successful e-tax adoption as a
specific e-government service in a developing country like Uganda, namely: What are the
requirements for successful e-tax service adoption by citizens in Uganda? How can the
proposed requirements be used to create a theoretical citizen-based model for e-tax
services adoption in Uganda? How can the proposed model be evaluated with e-tax
consumers in Uganda?
Adoption of e-tax services in Uganda: a model of citizen-based factors 261
This paper presents a model for e-tax adoption in Uganda as a specific e-government
service. Citizen-based factors for e-tax adoption in Uganda as identified in a field study
are used to develop the model. The factors identified are: benefits of e-tax adoption, trust,
attitudes, education, compatibility, awareness, accessibility, training, user support and
local language use. Theoretically, the model is underpinned by the technology
acceptance model (TAM) which it extends. The model contributes to the extension of
the existing e-government adoption models and can be applied to other developing
countries.
2 E-tax services
E-government consists of various fast moving fields, e-taxation being a very specific one
of them (Verlag, 2007). The author affirms that e-taxation means trans-organisational
processes with data transfer (upload and download) between the IT systems of the
professionals and those of the tax authorities; therefore, e-taxation should support tax
authority processes that include workflow systems and electronic record management on
the one hand, knowledge management and automated risk analysis to assess the
credibility of tax returns on the other hand. Electronic tax (e-tax) is described by MoICT
(2010) as the computerisation of tax operations including assessment, filing, processing
and payment procedures.
E-governance essentially means the electronic delivery of government services and in
the context of taxation this translates into online filing and assessment of tax returns;
electronic funds transfer to enable taxes to be paid and refunds to be issued across the
internet; different government departments sharing information on tax assessments to
support comprehensive policy decision making; web-based information portals for use in
educating taxpayers and researchers on taxation issues, and publishing information, hence
the overall aim of e-taxation is to replace cumbersome manual, bureaucratic service
systems with collaborative, efficient, process-driven and secure online delivery (Arya,
2011).
2.1 E-tax in Uganda

Uganda’s tax system is divided into central government and local government tax
structures. The principal taxes levied by the central government are income tax both on
individuals and companies; value added tax (VAT); import duty and excise duty,
all administered by the Uganda revenue authority (URA). Most taxes in Uganda are
charged on the basis of residential status and not on the basis of citizenship. The tax
payer is charged based upon the following factors: resident, resident but not ordinary
resident, non-resident (Focus on Uganda, 2009).
The Uganda Revenue Authority (URA) established in 1991 is charged with assessing
and collecting tax revenue specified, enforcing the laws regarding the collection of taxes
and accounting for the revenue collected to the central organisation (Uganda Revenue
Authority, 2012). The revenue department has boosted its tax collection by running an
integrated tax administration system that offers 24-h online services to taxpayers
(Morden, 2011). The system, called e-tax, which was installed in Uganda on 15th May,
2009, was initiated by URA to allow for the electronic mode of registrations, filings and
payments related to tax delivery and collections. The online option was started to shorten
the process of tax payment by avoiding queuing up of tax payers and to reduce costs such
as travel while filing tax returns. The e-tax system offers benefits to both the service
provider (URA) and the citizens. To the citizens, under e-tax and its e-filing and
e-payment features, the filing process takes 10 min compared to the old manual one that
took up to two days. Taxpayers have benefited under e-tax, through the efficient and
cost-free method of transaction it provides. It also reduced the amount of time to get a
tax identification number (TIN), which acts a taxpayer’s account in e-tax used to access
the URA’s online services and information. To the service provider, e-tax aids the URA
in processing income tax, VAT, local excise duty and other fees and licenses. E-tax
mirrors the procedures under the old system but it has however enhanced the old system
by putting all manual transactions online (Morden, 2011).
2.2 Challenges of e-tax adoption in Uganda

Much as the new e-tax system of payment in Uganda has become easy and simple,
it is still only used by the large companies. The small companies and individuals not
connected electronically are still finding it difficult to pay taxes (ESA-BMO Network,
2012). There are also inconsistencies in the working electronic systems. Much as many
users appreciate the electronic systems of the Uganda Revenue Authority, it still faces a
lot of problems:
• the online returns filing system of e-tax is prone to system failure and has little back
up support for customers (Harunah, 2011)
• low information technology penetration in Uganda
• insufficient taxpayer sensitisation for the online taxation system
• the system is in some cases too slow and time-wasting when faced with many
transactions (Uganda Radio Network, 2010; Kabafunzaki, 2010; Harunah, 2011)
• little user support especially during the busy dates of return filing
• lack of user training on the use of the e-tax system
• a negative attitude towards the e-tax system especially by the small business
operators who think that it is meant for big companies.
These factors have contributed to the slow adoption and use of e-tax systems leaving tax
payers stuck to submitting tax details manually. Taxpayers cite issues of the system being
complicated, difficult to use, lack of feedback, high internet costs coupled with an
internet dial-up system that slows communication with URA (Kabafunzaki, 2010;
Magumba, 2011). Taxpayers also cite the insistence by URA for customers to provide
manually printed back up returns as a disincentive for the adoption of the online e-tax
system (Magumba, 2011).
E-tax systems are part of the effort to adopt e-government in Uganda. E-tax adoption
initiatives therefore face similar constraints as the rest of the e-government adoption
efforts. In Uganda, despite the various e-government initiatives, several challenges
remain for ensuring implementation and adoption of e-government (MoICT, 2009).
These are:
• inadequate IT human resource base

• a high level of digital marginalisation (digital divide) especially in the rural areas
• inadequate research and development in IT to inform government’s uptake of
relevant applications
• inadequate legal and regulatory framework for the sector
• dependency on satellite bandwidth for the provision of services leading to high costs
of internet services
• high cost of IT equipment and software; inadequate IT integration into government
business processes leading to low utilisation during the delivery of various public
services
• lack of awareness and education; and gender differences (MoICT, 2009; Wougnet,
2009).
E-government has therefore not been fully realised in Uganda and there is inadequate
research and development in IT to inform government’s uptake of relevant applications
(MoICT, 2009). The IT adoption models as used in the developed countries cannot be
generalised for Uganda as a developing country. The models are developed to meet
the specific needs of individual countries; hence there is no one acceptable model for
e-government adoption globally.
The question therefore remains of what factors help to explain the gap between the
availability of e-tax services and its adoption and usage as a specific e-government
service in Uganda. Bridging this gap requires answers to the following specific research
questions: What are the requirements for successful e-tax service adoption by citizens in
Uganda? How can the proposed requirements be used to create a theoretical citizen-based
model for e-tax services adoption in Uganda? How can the proposed model be evaluated
with e-tax consumers in Uganda? This paper proposes a citizen-based model for e-tax
adoption in Uganda. The model extends an existing e-government adoption model that
considers citizen-based factors of trust, education, accessibility, attitudes, compatibility,
benefits, awareness, local language use, user support and training.
3 Related literature
E-government services are mostly provided using the infrastructure of ICT. Adoption of
e-government services therefore calls for an understanding of the related information
technology (IT) adoption models (Bwalya and Healy, 2010). The e-government adoption
models presented here and compared are the TAM (Davis, 1989); the e-governance
acceptance model (Sahu et al., 2004), a model for successful e-government adoption in
India (Saibal, 2007), the conceptual adoption model for e-government (Al-Adawi et al.,
2005; Bwalya and Healy, 2010; Kumar et al., 2007; Warkentin et al., 2002).
“The Technology Acceptance Model (TAM) (Davis, 1989; Davis and Arbor,
1989). TAM posits that users’ adoption of any new information system is
determined by the users’ intention to use that system, which is in turn
determined by the users’ beliefs about the system. TAMs main dependent
constructs are the behavioural intention to use and system usage. Its major
independent construct(s) are the Perceived Usefulness (PU) and Perceived Ease
of Use (PEOU) as shown in Figure 1.”
Figure 1 The technology acceptance model
Source: Davis (1989) and Venkatesh et al. (2003)
TAM has been tested empirically in different parts of the world, yielding statistically
reliable results and it has proved to be one of the most reliable and easy models of
explaining individual’s intention of adoption of a technology (Moon and Kim, 2001).
TAM has been used by many researchers especially in information system to achieve
a better understanding of IT adoption and its success in organisations (Carter and
Belanger, 2004, 2005). TAM has proven to be a strong and robust framework to clarify
adoption pattern of users (Horton et al., 2001). PU is the degree to which a person
believes that using a particular system would enhance his or her job performance. PEOU
is the degree to which a person believes that using a particular system would be free of
effort. TAM, however, is limited because it concentrates on two factors of PU and PEOU
and does not take into account the fact that the technology can be abandoned even after
it had been earlier accepted due to other factors deemed important by citizens for
e-government to fully succeed. TAM leaves out citizen-based factors of trust, attitudes,
benefits, education, awareness, accessibility, training, user support, compatibility and
local language. Such limitations impede the use of TAM to adopt e-government.
The models presented in this section remain silent on factors of trust, attitudes,
education, awareness, accessibility, training, user support, benefits, compatibility and
local language which are crucial for citizens in adopting e-government for a developing
country like Uganda. These models have also largely been used in the developed
countries and cannot easily be applied in developing countries without taking into
consideration the country’s specific conditions and context. The need therefore remains
for their customisation if they are to suit the different contexts of individual developing
countries (Zarei et al., 2008).
3.1 Modelling e-tax adoption with TAM: related work

The TAM has been used as a supporting theory in various studies to explain e-tax
adoption models (Yusniza and Azmi, 2010; Wang, 2002; Ramlah et al., 2010; Boone,
2012; Ahmet et al., 2011; Azmi and Bee, 2010; Cheng-Tsung et al., 2010; Ozgen and
Turan, 2007). In the following sections, the works related to the model are presented
and discussed.
Tax e-filing adoption in Malaysia: a conceptual model (Yusniza and Azmi, 2010). This
study presents that e-filing is one of the e-government services that is gaining importance
for public to perform their responsibility to the country through tax payment. Despite the
rapid adoption of e-filing in Malaysia, the e-filing system is still unreliable especially at
peak period which lead to high level of risk perceived by taxpayers. The research adopts
and extends the theory of acceptance model (TAM) to include performance risk as a facet
of perceived risk; this is discussed further in the research framework section. Based on
TAM, the study investigated the effect of perceived risk on the behavioural intention of
tax payers to use e-filing facility provided by the government and propose a conceptual
model to further understand the role of perceived risk in influencing citizen behaviour
throughout the adoption process (Yusniza and Azmi, 2010), as shown in Figure 2.
Figure 2 A conceptual model for e-filing adoption in Malaysia
Source: From Yusniza and Azmi (2010)

Adoption of electronic tax filing systems: an empirical study (Wang, 2002). This paper
discusses the factors affecting the adoption of electronic tax-filing systems. Using the
TAM as a theoretical framework, this study introduces ‘perceived credibility’ as a new
factor that reflects the user’s intrinsic belief in the electronic tax-filing systems
and examines the effect of computer self-efficacy on the intention to use an electronic
tax-filing system (Wang, 2002), as shown in the following Figure 3.
Figure 3 The adoption of electronic tax filing systems model
Source: From Wang (2002)
The findings of this study strongly support the appropriateness of using the extended
TAM to understand people’s intention to adopt the electronic tax-filing systems.
The significant effects of perceived usefulness, PEOU, and perceived credibility on
behavioural intention were observed, with PEOU exerting a stronger influence than either
perceived usefulness or perceived credibility (Wang, 2002).
An integrated model on online tax adoption in Malaysia (Ramlah et al., 2010). This study
argues that online tax system or e-filing has gained its popularity since it was launched in
2006 in Malaysia, but the obstacles remain concern taxpayers who are really anxious
about the technology, lacking of technical skills, or do not trust of any online transaction.
This has led to the investigation of the factors that might contribute to the taxpayers’
adoption of the online tax or e-filing system. To fulfil this aim, a model of online tax
adoption was developed by integrating the TAM theory, diffusion and innovation (DOI)
theory, perceived characteristics of innovating (PCI) and other relevant factors including
social influence, web-based service quality, trust of the internet, trust of the government
political self-efficacy and perceived risk. The result exhibits that trust of the internet and
trust of the government are moderately related to intention to use, trust of the internet and
perceived risk are also significantly related (Ramlah et al., 2010).
E-government and citizen adoption of innovations (Boone, 2012). This study focuses on
state revenue agencies and their electronic tax-filing mechanisms for individual income
taxes. It examines the impact of factors, including beliefs about how user-friendly the
system is, how beneficial using the system is, computer competence and service cost.
The additional variables are suggested in part by Davis’ TAM and the results of the
analyses showed that the TAM is an appropriate underlying framework for assessing the
factors influencing adoption of electronic filing of state tax returns and the theoretical
model was evaluated, the results showed no difference in satisfaction with online vs.
filing by paper. Demographic variables like income, education and ethnicity were not
significant factors affecting filing method. When controlling for demographic variables
only age and gender emerged as a significant factors for filing method with younger
males being more likely to file electronically. When controlling for demographic factors,
none emerged as significant factors influencing satisfaction (Boone, 2012).
Factorial validation of Turkish adapted TAM (Ahmet et al., 2011). In this study, Davis’s
(1989) TAM is used as a theoretical framework to extend and complement extra tax
officials by acceptance of technological components of tax office automation system in
Turkey. The results show that perceived usefulness significantly influenced PEOU,
PEOU was found to be significant in influencing attitudes and attitude was significant in
influencing behavioural intention (Ahmet et al., 2011).
The acceptance of the e-filing system by Malaysian taxpayers (Azmi and Bee, 2010).
This study investigates the factors that lead to the acceptance of e-filing among taxpayers
by using TAM and proposes a model consisting of three constructs: perceived usefulness,
PEOU and perceived risk. The results showed that PEOU and perceived usefulness have
significant positive effects on behavioural intention, perceived risk was also a significant
predictor of behavioural intention, there was a negative effect of PEOU on perceived risk
(Azmi and Bee, 2010), as shown in Figure 4.
An empirical study of online tax filing acceptance model (Cheng-Tsung et al., 2010).
This study integrated Theory of Planning Behaviour (TPB) and TAM to investigate
the determinants affecting online taxpayers. The empirical results showed that attitude
was the primary factor affecting online tax filing, but attitude was also affected
by perceived usefulness, PEOU, tax equity, social norm and moral norm; both TAM
and TPB could be successfully integrated to explain online tax filing behaviours
(Cheng-Tsung et al., 2010).
Usage and adoption of online tax filing and payment system in tax management (Ozgen
and Turan, 2007). This study investigates e-declaration and e-tax payment systems in
Turkey. E-tax payment system (or online tax payment system, e-tax lodgement system) is
an information system by which taxpayers would pay electronically the tax due. The
paper makes an empirical assessment of adoption of e-tax filing system in Turkey with
TAM (Ozgen and Turan, 2007).
From the related work above, it is clear that the TAM by Davis (1989) has been used by
many researchers in e-tax adoption. However, it is evident that the models do not address
the citizen factors as identified in the fore literature.
Figure 4 The simplified model for the acceptance of the e-filing system by Malaysian taxpayers
Source: From Azmi and Bee (2010)
4 Deriving a model for e-tax service adoption in Uganda
In this section, the steps used to develop an e-tax adoption model for Uganda are
outlined. Requirements for the model as elicited in a field study are used to extend
the TAM. The following sections therefore present the model and its synthesis from the
requirements, supported by the theory of TAM.
4.1 The field study

A descriptive field survey using a questionnaire was undertaken to determine the
requirements for the model using consumers of e-tax services in Kampala Central area.
The target population was citizens (business persons, employees of government and
non-government organisations). An interview with the commissioner for domestic taxes
(CDT) of URA revealed that statistically, the number of e-tax users across the country as
of 31st May, 2011 with new TINs was 54,962. In Kampala, there were 40,648 e-tax users
spread across the four parts of the city (Kampala central, south, north and east) and in
Kampala Central, the population of e-tax users was estimated to be 8000. The sample size
needed was a function of the confidence interval of (+/−) 5%, a confidence level of 95%,
and the population size of 8000 competent respondents from Kampala Central.
The sample size was determined using the following formula (Krejcie and Morgan, 1970).
In equation (1)
Z 2 X (1 − X )
SS = (1)
C2
where SS is the required sample size; Z is the Z value (e.g., 1.96 for 95% confidence
interval); X is the percentage picking a choice expressed as decimal (0.5 used for sample
size needed), C is the confidence interval expressed as decimal (0.05 ± used for sample
size needed). Assuming a set alpha level at 0.05 and a level of acceptable error at 5%,
and an estimated standard deviation of the scale as 0.5, the sample size according to
equation (2) becomes:
(1.92)2 × (0.5) × (0.5)
SS = . (2)
(0.05) 2
The above helps to determine the required sample regardless of the size of the population.
The SS (required sample) indicates the sample that is expected to be considered with
reference to the confidence level of 95% and standard error of 5% as indicated in the
equation above. The 0.5 as value of ‘X’ indicates that there is a simple random technique
being used as a method of picking data from the population.
Considering the formula below, the new sample which shall hereby be referred
to as the actual sample for the selected population size shall be selected. The formula is:
New and actual sample = {SS/[1 + [(SS − 1)/P]]} where SS refers to the original sample
that does not have any reference to population size (384) and P refers to the size of the
population from which the sample is supposed to be taken (8000). Considering the
population at hand, the sample below can be obtained as:
{384/[1 + [(384 − 1)/8000]]} = 366.5. (3)
Of the 380 questionnaires that were given out to e-tax service users in Kampala Central,
292 of the respondents returned valid questionnaires. This presents a response rate of
79.5% (292/367 × 100 = 79.5%), approximately 80%. Probability sampling method was
used where every unit in the population had a chance of being selected in the sample
using simple random sampling. This method gave equal chances to all respondents
identified in the study. Interviews were used to gain information on requirements that
were later confirmed by data collected using questionnaires. Data was then collected
using semi-structured close-ended questionnaires to determine the requirements for
the citizen-based e-tax adoption model in Uganda. Data collected was categorised,
quantified, coded and arranged in themes with respect to the objectives of the study. The
questionnaire was checked for reliability and validity.
Mixed method research was used in this study. Mixed research is a general type
of research in which quantitative and qualitative methods, techniques or other paradigm
characteristics are mixed in one overall study, Tashakkori and Teddlie (2003). A mixed
method research design as suggested by Woodgate and Wilkins (2008) was followed in
this study. This study adopted the pragmatic research philosophy as the primary
philosophy of mixed research is that of pragmatism (Burke et al., 2007). Pragmatism
allows mixing data collection methods and analysis procedures within the research
process to suit the research questions. Researchers can use the approach or mixture of
approaches that works the best in a real-world situation. Pragmatism supports using both
quantitative and qualitative methods and provides several advantages over mono-method
approaches (Creswell, 2003; Greene, 2007).
4.2 Reliability of the questionnaire

Reliability is the extent to which results are consistent over time. If the results of a study
can be reproduced under a similar methodology, then the instrument is considered
reliable (Joppe, 2000). Constructs in the study were examined for reliability using the
Cronbach’s alpha (Cronbach, 1970) and the results are as presented in Table 1.
Table 1 shows that all constructs met the required level of Cronbach’s alpha as the
values were above 0.70 – the acceptable level. All values were above 0.80 and are
considered very well (Nunnally, 1978). The questionnaire can therefore be relied upon.
Table 1 Summary of reliability of scales measures
Construct Cronbach’s alpha

E-government adoption benefits 0.865
Trust 0.853
Attitudes towards use of e-government services 0.874
Factors limiting use of e-government services 0.905
Use of e-taxation services in Uganda 0.963
Reasons for not using the e-tax services 0.878
4.3 Validity of the constructs

Validity determines whether the research truly measures that which it was intended to
measure (Muhammad et al., 2008). Factor analysis was used to examine convergent and
discriminant validity. Factor analysis is statistically used to determine the correlation
among variables in a dataset and provides a structure to group variables based on strong
correlations helping to detect misfit variables. Principal component analysis was used
with Varimax Kaiser normalisation as it considers all the available variance. Convergent
validity indicates the degree to which the items measuring the same construct are
correlated. Discriminant validity indicates the degree to which the variables on the factors
are distinct and uncorrelated. Low values indicate candidates for removal after examining
the pattern matrix.
Validity on reasons for not using e-tax services – Respondents’ reasons for not using
some of the e-tax services in Uganda were also evaluated for discriminant and convergent
validity and the results are presented in Table 2.
Table 2 on the component factor loadings indicates that the factors loaded highly on
their own factors with two components extracted. Two factors, with an asterisk (*) had
cross loading, that is, item 4 and item 12 but these items loaded higher on the constructs
they measured. Discriminant validity was therefore achieved. Convergent validity on the
reasons for not using the e-tax services was significant because all the factors after
extraction loaded higher than 0.50. The results indicated that two factors were extracted
with an eigenvalue above one (1.0). All the factors loaded higher than 0.50 implying that
convergent validity was desirable and therefore achieved.
Table 2 Component factor loading on the reasons for not using e-tax services
Component
Reasons for not using e-tax services 1 2
Lack of benefits in using e-tax services 0.836
Lack of trust in the e-tax system 0.827
Negative attitudes towards the e-tax system 0.784
Lack of computer/internet training 0.594* 0.555
Unreliable power supply 0.612 0.418
Limited access to a computer 0.707
Limited access to the internet 0.784
Inadequate training in using the e-tax system 0.854
Limited user-support from e-tax service providers 0.837
Unreliable e-tax service (constant breakdown) 0.755
Lack of awareness of the e-tax system 0.741
Use of non-local language (English) content in the e-tax system 0.675* 0.430
Difficulty in using the e-tax system 0.681
Extraction method: Principal Component Analysis.
Rotation method: Varimax with Kaiser Normalisation.
4.4 Field study results

The findings presented in this section are results from the data collected through
interviews and questionnaires issued to the citizens of e-tax services in Kampala (central
Uganda).
Use of e-tax services in Uganda: Data was collected on the use of different e-tax services
in Uganda and the findings are presented in Figure 5.
Figure 5 shows that the most used e-tax service is the e-forms and templates for
registration (72%) and the least e-tax service used is the e-tax query (68%).
Reasons for not using the e-tax services: The study also elicited responses on a number of
attributes proposed as reasons why respondents did not use some of the e-tax services and
the responses are presented in Figure 6.
Figure 6 reveals that for most respondents, reasons for not using e-tax services were:
lack of perceived benefits (74%); lack of trust (73%); negative attitudes towards the e-tax
system (70%); lack of computer and internet training (62%); limited access to computers
(58%); limited internet access (54%); inadequate training about e-tax services and user
support (55); unreliable e-tax service (60%); lack of awareness of the e-tax system
(59%); use of non-local language (62%); difficulty in using the e-tax system (60%).
Figure 5 Use of e-tax services in Uganda (see online version for colours)
Figure 6 Reasons for not using e-tax services in Uganda (see online version for colours)
4.5 A citizen-based e-tax adoption model

The model proposed in Figure 7 extends TAM (Davis, 1989) using requirements obtained
from the field study. Apart from the four factors established in TAM: perceived
usefulness, PEOU, behavioural intention to use and actual system usage, the new model
adds 10 citizen factors as theoretical constructs. These factors are explained under
usability and usage. Nielsen (2012) defines usability as a quality attribute that assesses
how easy user interfaces are to use. According to Nigel (1995), usability is closely related
to ease of use and Davis (1989) defined PEOU as the degree to which a person believes
that using a particular system would be free of effort. Usability factors that enhance
PEOU in this study include awareness, accessibility, compatibility and local language.
Usage factors that enhance perceived usefulness are trust, training, user support, attitudes,
education and benefits as shown in Figure 7.
Figure 7 outlines the existing model by Davis (1989) and Venkatesh et al. (2003) and
presents new features that are useful for adopting e-tax in the Ugandan context.
Figure 7 The citizen-based e-tax adoption model (extension of TAM)
The findings point to lack of awareness as limiting e-tax adoption in Uganda. Awareness
is the state or ability to perceive, to feel or to be conscious of events and objects. In the
early stages of implementing IT, awareness contributes to a willingness to adopt new
technologies (Jaruwachirathanakul and Fink, 2005). For ordinary citizens to recognise the
value of the new technology, its potential utility must be known. If potential users are not
aware of online services, they will not be able to use them (Common Wealth of Australia,
2011). Sensitisation programs need to be created in the traditional media (radio and
newspaper columns), television and government websites to create awareness about e-tax
in Uganda targeting various e-tax users.
Accessibility is “the degree to which web information is accessible to all human
beings and automatic tools”. It influences the citizen’s experience with websites and their
satisfaction and adoption of new technology. Accessibility is an important method of
building useful user-centred e-government services (Abanumy et al., 2005). For e-tax
services to be adopted in Uganda, citizens should access computers, the internet, training,
user support and specific e-government service software. To increase on citizen access to
computers and internet, the government of Uganda could exempt all computers from
excise duty and reduce internet access charges. Communal access in rural areas can be
provided through village computer centres or kiosks with aides who can train users in
basic computer skills and in using the e-tax system, including those physically disabled.
Compatibility is a significant determinant in citizen’s intention to use e-government

services (Carter and Belanger, 2004, 2005). An innovation is more likely to be adopted
when it is compatible with the individual’s job responsibilities and value system (United
Nations E-Government Survey, 2008). Therefore, e-tax services should be compatible
with existing values, experiences and beliefs of the citizens.
Language of content: Yong (2004) stated that “simply providing convenient or
low-cost access will not entice citizens to go onto the internet if there is little online
content in the local language”. To encourage citizens to take up e-government services,
governments need to do their part to promote local industries that develop online content
in the local languages. The main target for e-tax adoption should be the potential users
with little or no formal education. Local language content should be tailored to different
communities. The government could also find ways of promoting local industries and
government departments that develop online content in the local language.
Training involves identifying the e-tax service training needs (needs assessment),
setting objectives for the training, design the training program (means and methods to be
used), identifying users who need the training, conducting the training with follow-up,
observing users for signs of adoption or rejection (with feedback) and comparing training
outcomes with objectives. To increase literacy among e-tax users in Uganda, mass
trainings could be organised and delivered both in English and local languages. Local
tele-centres could also be established and used to train users at a fee or training could be
offered at all URA offices country-wide to citizens time-tabled per sub-county.
Trust: Government-citizens’ trust is an important catalyst of e-government adoption
(Al-Adawi et al., 2005). Trust is an expectation that alleviates the fear that one’s
exchange partner will act opportunistically, it is a belief that others will behave in a
predictable manner (Senne, 2005). Citizens must have a strong trust in the security of
electronic communications in order for e-government to succeed and prosper (Al Sawafi,
2003). The main concern of citizens in many countries around the world is the fear that
their information could be misused; therefore, trust is a key element for successful
e-government adoption (Layne and Lee, 2001). Trust factors need to be addressed in
Uganda by keeping lines of communication open between URA and e-tax users.
An increase in the level of trust for e-tax systems should result into increased cooperation
and information sharing between citizens and URA.
Attitudes are constructs that represent an individual’s degree of like or dislike for
something. Attitudes are generally positive or negative views of a person, place, thing or
event (Schwitzgebel, 2006). Some citizens may have a negative attitude towards
electronic services and that they would prefer to stay with traditional methods, which
for most is the paper-based way and in this case attitudes towards using government
e-services represent one of the main barriers for e-government adoption (Vassilakis et al.,
2005). The negative attitudes that need to be addressed are: e-tax services are prone to
error, being difficult to use, confidentiality of information, issues of fraud, paper-based
systems give better control over information, e-government systems do not add any
benefit/value to users, men are more computer/internet literate than women.
User support – A dedicated and appropriate user support mechanism may assure
individual citizens of appropriateness of engaging in e-government and this will
positively impact on both e-government adoption and continuance use of e-government
(Bwalya and Healy, 2010).
The benefits of e-tax adoption: Kamal and Themistocleous (2006) stated that from an
African perspective, the intention to engage in e-government is also partially influenced
by the perceived benefit of using such a platform. This perceived benefit can be looked
at as a return on investment (ROI) of one’s time, effort, financial investment,
and psychological preparedness to seek technology as a platform for engaging with
government. If the return on such basic investment is low, it is anticipated that an
ordinary African will go for the traditional way of interaction with the government.
Education is any act or experience that has a formative effect on the mind, character
or physical ability of an individual. Three key elements of education should be
considered for successful adoption for any technology: awareness of the internet;
understanding of the internet and workers with information technology skills
(Schwitzgebel, 2006). The most frequent use of e-government information and services
comes from populations who are experienced in using the internet as a technology.
Education is one of the problems related to e-government adoption and as citizens’
education rises, their knowledge in using the internet increases (Kumar et al., 2007).
5 Model evaluation
Evaluation is the procedure of checking whether the model is applicable or not.

To evaluate this model, a group of 120 e-tax citizens and especially accountants in
Kampala Central were availed with the model and asked to rank it according to a
basis on the validation questionnaire. Out of the 120 questionnaires distributed,
100 were returned completed and deemed usable yielding an effective response rate of
83% (100/120 × 100).
5.1 Reliability of validation questionnaire

The questionnaire was examined for reliability using Cronbach’s alpha values.
The results as presented in Table 3 indicate that all values approximate 0.70 and above,
which is the acceptable range (Nunnally, 1978). Most values were above 0.80 which is
considered very good, except for training (0.676). The questionnaire was thus ascertained
as reliable.
Table 3 Reliability statistics of the validation questionnaire
Variable Cronbach’s alpha No. of items

PU 0.907 3
PEOU 0.879 2
Awareness 0.889 3
Accessibility 0.759 3
Training 0.676 3
Trust 0.859 3
Attitudes 0.873 3
Education 0.928 2
Benefits 0.909 3
Local language 0.871 2
User support 0.916 3
5.2 Regression analysis results for the e-tax service adoption model
Regression analysis was used to explain the relationship between perceived usefulness
(dependent variable) and training, trust, attitudes, user support, benefits and education
(independent variables) in one set; PEOU (dependent variable) and awareness, access,
compatibility and local language (independent variables) in another set. To explain
perceived usefulness and PEOU, multiple regression analysis was performed between the
dependent and independent variables. Multiple regression is a technique that allows a
number of factors to enter the analysis separately so that the effect of each can be
estimated. It is valuable for quantifying the impact of various simultaneous influences
upon a single dependent variable. The extended TAM explained up to 40.3% of variance
in PEOU and 89.2% of variance in perceived usefulness is presented in the model
summarised in Tables 4 and 5.
Table 4 Summary of the multiple regression analysis with perceived ease of use as dependent
variable
Standardised
Unstandardised coefficients coefficients
B Std. error Beta t Sig.
(Constant) 0.034 0.471 0.072 0.943
Accessibility 0.741 0.153 0.547 4.835 0.000
Awareness 0.217 0.149 0.169 1.452 0.150
Local language –0.172 0.106 –0.178 –1.618 0.109
Compatibility 0.114 0.120 0.097 0.948 0.346
a. Dependent Variable: Perceived Ease of Use.
R = 0.634, R2 = 0.403, Adjusted R2 = 0.377.
Table 5 Summary of the multiple regression analysis with perceived usefulness as dependent
variable
Standardised
Unstandardised coefficients coefficients
Variables B Std. error Beta t Sig.
(Constant) –0.622 0.167 –3.719 0.000
Training 0.536 0.073 0.453 7.317 0.000
Trust –0.030 0.050 –0.031 –0.612 0.542
Attitudes 0.417 0.079 0.365 5.305 0.000
User support –0.045 0.057 –0.044 –0.784 0.435
Benefits 0.092 0.074 0.089 1.242 0.217
Education 0.184 0.066 0.180 2.811 0.006
a. Dependent Variable: Perceived Usefulness.
R = 0.944, R2 = 0.892, Adjusted R2 = 0.885.
In Tables 4 and 5, R is a measure of the correlation between the observed value and the
predicted value of the dependent variable. R2 is the amount of variation in the dependent
variable that is accounted for by variation in the independent variables. It is measured by

the value of the coefficient of determination, often called R2 adjusted because R2 tends to
over-estimate the success of the model when applied to the real world, so an adjusted R2
value is calculated which takes into account the number of variables in the model and the
number of observations (participants) a model is based on. If the adjusted R2 value is
closer to 1, then the result would be better, because the adjusted R2 value gives the most
useful measure of the success of a model. If R2 adjusted is 1, then the regression model is
accounting for all the variation in the dependent variable. In Table 4, the adjusted R2 is
0.377 which means the independent variables account for 37.7% of PEOU. In Table 5,
the adjusted R2 is 0.885 which means the independent variables account for 88.5% of
perceived usefulness. This figure is closer to 1 meaning the model can be applied
successfully.
To explain perceived usefulness and PEOU, multiple regression analysis was
performed between the dependent variables and independent variables and the results are
presented in Tables 4 and 5.
In Tables 4 and 5, β coefficients (the plus or minus sign) help to interpret the
direction of the relationship between variables. If a β coefficient is positive, then the
relationship of this variable with the dependent variable is positive and if the β coefficient
is negative then the relationship is negative, if the β coefficient is equal to 0 then there is
no relationship between the variables. In Table 4, the independent variables that have
positive relationships with PEOU are awareness, accessibility and compatibility and the
one with negative relationship is local language. In Table 5, the independent variables
that have positive relationship with perceived usefulness are training, attitudes, benefits
and education; those with negative relationships are trust and user support. Standard
error, which is the distance between the line and all the points, indicates whether the
regression analysis has captured a relationship that is strong or weak. The closer a line is
to the data points, overall, the stronger the relationship.
The beta value is a measure of how strongly each independent variable influences
the dependent variable. The beta is measured in units of standard deviation. The Beta
value is used to assess the strength of the relationship between each independent variable
to the dependent variable and the higher the beta value the greater the impact of the
independent variable on the dependent variable. The independent variable with higher
beta values in Table 4 is accessibility (β = 0.547) and the independent variables in
Table 5 with higher beta values are training (β = 0.453), attitudes (β = 0.365) and
education (β = 0.180). The tolerance values (T) are a measure of the correlation between
the independent variables and can vary between 0 and 1. The closer to zero the tolerance
value is for a variable, the stronger the relationship between this and the other
independent variables.
The significance value (P value) gives the impact of each independent variable on the
dependent variable. A smaller P value suggests that the independent variable is having a
significant impact on the dependent variable. Tables 4 and 5 present the results of the
multiple regression showing the influence of the independent variables on perceived
usefulness and PEOU (the dependent variables) at a significance level of 0.05 and for
independent variables to have significant influence on the dependent variables, their
significance value should be 0.05 and below. The summary of the multiple regression
analysis shows the values of coefficients and the direction of relationship with the
dependent variable including their respective P-values (significance values).
Evidence derived from Table 4 shows that out of the four independent variables, only
one was found to impact a significant influence on PEOU usefulness. This is because
its P-value was less than the level of significance (P < 0.05) and according to the
results, this variable is accessibility (P = 0.000). This suggests that holding other factors
constant, PEOU in the context of this study is dependent on accessibility, other variables
playing a negligible role. In Table 5, out of the six independent variables, three are found
to exert a significant influence on perceived usefulness as their respective P-values are
less than the level of significance (P < 0.05). These variables are training (P = 0.000),
attitudes (P = 0.000) and education (P = 0.006). These results suggest that holding other
factors constant, perceived usefulness is dependent on the three independent variables,
others playing a negligible role.
The following factors were not significant at the 0.05 level in Table 4: awareness
(beta 0.169 and significance level 0.150), compatibility (beta 0.097 and significance level
0.346), local language (beta −0.178, significance level 0.109) and in Table 5: trust (beta
−0.031, significance level 0.542), user support (beta −0.044 and significance level 0.435),
benefits (beta 0.089, significance level 0.217). This is because their significant levels are
above the 0.05 significance level in this study. The factors determining perceived
usefulness were training (β = 0.453, P < 0.000), attitudes (β = 0.365, P < 0.000),
education (β = 0.180, P < 0.006) and the factor determining PEOU is accessibility
(β = 0.547, P < 0.000). Trust, user support, benefits, awareness, compatibility and local
language were not significant at the 0.05 level in this model as shown in Figure 8.
Figure 8 Results of the regression analysis (significant relationships at p < 0.05)
In Figure 8, the results of the regression analysis in Tables 4 and 5 are mapped on the
e-tax adoption model highlighting the factors that have significant impact on perceived
usefulness and PEOU. The results indicate that training, attitudes and education
significantly impact perceived usefulness and accessibility impacts PEOU. The model
explained up to 89.2% of the variance in perceived usefulness and up to 40.3%
of the variance in PEOU. This means that the extended model is adequate and can be
used for e-tax adoption in Uganda.
With respect to training, the results showed that training had significant influence on
perceived usefulness implying that in practice, training of the e-tax citizens in e-tax
system use would make them perceive e-tax as useful leading to adoption of e-tax and
must be taken into account while implementing e-tax services. Regarding attitudes, the
results showed that it had a significant effect on perceived usefulness. This means that if
the negative attitudes of the e-tax consumers are addressed, they will perceive e-tax as
useful. Education equally has a significant effect on perceived usefulness. This means
that if the e-tax consumers are educated in information technology skills and internet use,
they will perceive e-tax services as useful.
The results showed that accessibility has a direct significant effect on PEOU. This
means that if the e-tax consumers have access to computers, internet, training, user
support among others, e-tax services will be perceived as easy to use.
Results from the multiple regression analysis show that trust, user support and
benefits have no significant effect on perceived usefulness; compatibility, awareness and
local language also have no significant effect on PEOU.
5.3 Implications of the research

The outputs of this research present several far reaching theoretical and practical
implications for e-tax service providers and the consumers.
First of all, the results contribute to the development of a customised model for
successful e-tax adoption in Uganda. This provides URA department with a new toolkit
for adoption of e-tax services and guides them to be more responsive to citizen-based
e-tax adoption factors as an option to increase the adoption rates. It will therefore be vital
for this department to put these factors at the forefront and deter from dissipating
financial resources on e-tax projects that may eventually not bring benefits because of
low adoption rates.
To the citizens and especially e-tax service users, the model addresses the
demand-side perspective of e-tax by identifying and considering the factors that affect
them. This is hoped to enhance adoption of e-tax services as these crucial factors
should be addressed first before the service is rolled out. The model is generic and
can be applied to other e-government services like e-health, e-agriculture, e-voting,
e-procurement that target the same citizens.
To the researcher community, this research adds to the existing e-tax literatures by
focusing on the citizen-based factors for e-tax that addresses the demand-side perspective
of e-tax adoption. Other researchers and scholars will be in position to refer to the
contents of the study for any future related studies.
For other transitioning countries – Other transitioning countries that have
implemented e-tax but are faced with low adoption rates would use the designed model to
ensure successful adoption of e-tax services in their countries since conditions in
transitioning countries are almost the same. This can help the countries deter from
dissipating financial resources on e-tax projects that may eventually not bring benefits
because of low adoption rates.
5.4 How to improve relatively low use of e-tax services

The study established the citizen-based factors for successful e-tax adoption in Uganda
and therefore recommends the following for improving the relatively low use of e-tax
services:
Training of e-tax service consumers should be highly effected throughout the
implementation of e-tax services and this should be done before, during and after the
implementation of any e-tax endeavour to subdue the possibility of its failure and
increase benefits from adoption and usage of the services. The research found that
training had a significant effect on perceived usefulness; therefore, government
departments intending to offer e-services should address the training needs of the
consumers (especially training in the specific e-government service/software) and
provide continuous user support if they want those services to be adopted successfully.
Implementation of e-tax services should focus on changing the negative attitudes of
consumers towards e-tax services. Results revealed that respondents’ negative attitudes
towards e-tax deterred them from using the services; also, validation results showed that
training had a significant effect on perceived usefulness. This research therefore
recommends that training and creation of awareness be undertaken to reduce on such
negative attitudes before implementing e-tax and other e-government services as this is
one of the main barriers for e-tax adoption.
Much as consumers may be aware of the e-tax services and their benefits, if they do
not have access to computers, internet and training, usage of these e-services will still
remain low. The e-tax service provider and organisations need to ensure that citizens
have skills in using the internet, and generally IT skills and while implementing these
services, trust factors of integrity, reliable network, authentication and availability are
considered to remove the fear in consumers that their information could be misused.
In the process of sensitising e-tax consumers, both English and local languages should be
used to enable those who do not know English benefit. This will increase the usage
of e-tax service and in turn increase adoption rates as benefits are enjoyed, results from
this study revealed that use of local language in e-tax services is important for successful
adoption.
E-tax service providers should also consider important the issue of compatibility of
the e-tax services with the job responsibilities of consumers for a successful adoption.
Finally, a policy for e-tax adoption needs to be formulated and implemented that will
address both supply-side and demand-side issues. This can generally guide e-government
service providers in implementation of the e-services and ensure successful adoption.
6 Conclusions and future work
The model presented in this study outlines the requirements for successful e-tax adoption
as awareness, accessibility, training, user support, local language, trust, attitudes,
benefits, education and compatibility. These requirements are used to extend the TAM
(Davis, 1989; Venkatesh et al., 2003). Theoretically, the model provides the basis for
adoption of e-tax services in Uganda. The model contributes to e-tax adoption literature
by focusing specifically on the citizen-based factors. The model is generic and can be
applied to other developing countries that have similar contexts as Uganda.
A regression analysis provided support for relating the citizen-based factors

(awareness, accessibility, attitudes training, education) to the existing TAM construct of
perceived usefulness. Whereas TAM has been successfully used to predict technology
adoption in other domains (health and education), this model provides support for using
TAM as a theory to predict e-tax adoption for a developing country like Uganda. Future
work remains to implement and validate this model.
Literature revealed that most governments have focused on the supply side
of e-government adoption; therefore, based on the results of this study, the directions for
future research to better understand and enhance e-tax adoption and use are as follows:
The extended version of TAM needs to be modified and in developing this new
model, new constructs need to be tested as independent variables to TAM and
independent variables need to be identified and tested especially for PEOU. There is a
need to carry out additional research to test the extended TAM with different e-tax
services covering a wider scope (both in rural and urban settings) and the results
compared. This research also proposes that further research be carried out on the level of
adoption of e-tax services in Uganda both in rural and urban settings. The results can be
used to devise specific strategies for increasing adoption both in rural and urban areas.
The extended TAM needs to be evaluated including the variables in the original TAM to
find out the impact of the external variables on TAM’s perceived usefulness, the impact
of perceived usefulness on behavioural intention to use, actual system use and e-tax
adoption. This research proposes the adoption process to be studied in future to
understand the difference stages, activities and responsible persons.
References
Abanumy, A., Al-Badi, A. and Mayhew, P. (2005) ‘e-government website accessibility: in-depth
evaluation of Saudi Arabia and Oman’, The Electronic Journal of e-Government, Vol. 3,
No. 3, pp.99–106, Available online at www.ejeg.com
Ahmet, F.C., Serkan, B. and Tamer, B. (2011) ‘The acceptance of tax office automation system
(Vedop) by employees: factorial validation of Turkish adapted technology acceptance model
(TAM)’, International Journal of Economics Finance, Vol. 3, No. 6, pp.107–116.
Al Sawafi, A. (2003) ‘E-governance technologies for enabling trust in citizen relation
management’, Proceedings of the Symposium on E-Government: Opportunities & Challenges,
10–12 May, Muscat, Oman.
Al-Adawi, Ζ., Yousafza, S. and Pallister, J. (2005) ‘Conceptual model of E-government’,
Presented at the Second International Conference on Innovations in Information Technology
(IIT’05), 26–28 September, Dubai, UAE, pp.1–10.
Arslan, A. (2009) ‘Cross-cultural analysis of European e-government adoption’, World Applied
Sciences Journal, Vol. 7, No. 9, pp.1124–1130.
Arya, M. (2011) Auditing E-Taxation: Issues Challenges, p.35, Accessed on 8th February, 2012,
Http://Www.Intosaiitaudit.Org/
Azmi, A.C. and Bee, N.L. (2010) ‘The acceptance of the e-filing system by Malaysian taxpayers: a
simplified model’, Electronic Journal of e-Government, Vol. 8, No. 1, pp.13–22.
Bertot, J.C., Jaeger, P.T. and Mcclure, C.R. (2008) ‘Citizen-centered e-government services:
benefits, costs, research needs’, The Proceedings of the 9th Annual International Digital
Government Research Conference, 18–21 May, Montreal, Canada, pp.137–142.
Boone, M.A. (2012) E-Government Citizen Adoption of Innovations: Factors Underlying Citizen
Use of the Internet for State Tax Filing, Accessed on 1st February, 2013, Http://
Gradworks.Umi.Com
Burke, R.J., Anthony, J.O. and Lisa, A.T. (2007) ‘Toward a definition of mixed methods research’,
Journal of Mixed Methods Research, Vol. 1, p.112.
Bwalya, K.J. and Healy, M. (2010) ‘Harnessing e-government adoption in the Sadc region:
a conceptual underpinning’, Electronic Journal of E-Government, Vol. 8, No. 1, pp.23–32.
Carter, L. and Belanger, F. (2004) ‘Citizen adoption of electronic government initiatives’,
Proceedings of the 37th Hawaii International Conference on System Sciences, Hawaii,
5–8 January, IEEE, USA.
Carter, L. and Belanger, F. (2005) ‘The utilization of e-government services: citizen trust,
innovation acceptance factors’, Information Systems Journal, Vol. 15, No. 1, pp.5–25.
Cheng-Tsung, L., Shaio-Yan, H. and Pang-Yen, L. (2010) ‘An empirical study of on-line tax filing
acceptance model: integrating TAM and TPB’, African Journal of Business Management, Vol.
4, No. 5, pp.800–810.
Common Wealth of Australia (2011) What are the Benefits of Using E-Tax?, Accessed on
6th February, 2012, Http://Www.Tax-News.Com/News/
Creswell, J. (2003) Research Design. Qualitative, Quantitative Mixed Methods Approaches,
2nd ed., Sage, Thousand Oaks.
Cronbach, L. (1970) Essentials of Psychology Testing, Harper Row, New York, USA.
Davis, F.D. (1989) ‘Perceived usefulness, perceived ease of use user acceptance of information
technology’, MIS Quarterly, Vol. 13, No. 3, pp.318–340.
Davis, F.D. and Arbor, A. (1989) ‘Perceived usefulness, perceived ease of use, user acceptance of
information technology’, MIS Quarterly, Vol. 13, No. 3, September, pp.319–340.
ESA-BMO Network (2012) Accessed on 28th January, 2012, Http://Www.Esabmonetwork.Org/
Focus on Uganda (2009) The Tax System, Accessed on 8th February, 2012, Http://
BusinessinUganda.Com/Taxes
Greene, J.C. (2007) Mixed Methods in Social Inquiry, John Wiley & Sons, San Francisco.
Harunah, L. (2011) Taxpayers Cry Foul Over E-Tax System, the Sunrise, Accessed on 28 January,
2012, http://Www.Sunrise.Ug/Taxpayers-Cry-Foul-Over-E-Tax-System.Html
Horton, R., Buck, T., Waterson, P. and Clegg, C. (2001) ‘Explaining intranet use with technology
acceptance model’, Journal of Information Technology, Vol. 16, pp.237–249.
Infodev (2002) The E-Government Handbook for Developing Countries, Center for Democracy &
Technology, Worldbank, Washington DC, Available: www.infodev.org/en/Document.16.pdf
Jaruwachirathanakul, B. and Fink, D. (2005) ‘Internet banking adoption strategies for a developing
country: the case of Thailand’, Internet Research: Electronic Networking Applications Policy,
Vol. 15, No. 3, pp.295–311.
Joppe, M. (2000) The Research Process, Accessed 15th October, 2011, From Http://Www.
Ryerson.Ca/
Kabafunzaki, H. (2010) Taxpayers Cry Out on E-Tax, Accessed Http: //Www.Newvision.Co.Ug/
Kamal, M.M. and Themistocleous, M. (2006) ‘A conceptual model for EAI adoption in an
egovernment environment’, European Mediterranean Conference on Information Systems
(EMCIS) 2006, 6–7 July, Costa Blanca, Alicante, Spain, pp.1–11.
Krejcie, R.V. and Morgan, D.W. (1970) ‘Determining sample size for research activities’,
Educational Psychological Measurement, Vol. 30, pp.607–610.
Kumar, V., Mukerji, B., Irfan, B. and Ajax, P. (2007) ‘Factors for successful e-government
adoption: a conceptual framework’, The Electronic Journal of E-Government, Vol. 5, No. 1,
pp.63–77.
Layne, K. and Lee, J. (2001) ‘Developing fully functional e-government: a four stage model’,
Government Information Quarterly, p.122.
Magumba, T. (2011) Ura E-Tax System Fails, Accessed on 30th January, 2012. Http://
Www.Razor.Ug/
Ministry of Information Communications Technology (MoICT) (2009) Ministerial Policy

Statement, Vote: 020, Financial Year 2009/2010.
Ministry of Information Communications Technology (MoICT) (2010) National Electronic
Government (E-Government) Framework (Draft Final), p.24.
Moon, J.W. and Kim, Y.G. (2001) ‘Extending the TAM for a world-wide-web context’,
Information & Management, Vol. 38, No. 4, pp.217–230.
Morden, J. (2011) E-Tax System Brings Savings to Uganda Taxman, Accessed on 27th March,
2011, Http://Www.Futuregov./E-Tax-System-Brings-Savings-Uganda-Taxman/
Muhammad, B., Muhammad, T.A. and Muhammad, A. (2008) ‘Reliability validity of qualitative
operational research paradigm’, Pak. J. Stat. Oper. Res., Vol. IV, No. 1, pp.35–45.
Nielsen, J. (2012) Usability 101, Accessed on 12th November, 2012, Http://Www.Useit.Com
Nigel, B. (1995) ‘Usability is quality of use’, Yokohama, Anzai, Y and Ogawa, K. (Eds.):
Proceedings of the 6th International Conference on Human Computer Interaction, Elsevier,
Available: http://www.usabilitynet.org/bevan/publications.htm
Nunnally, J.C. (1978) Psychometric Theory, 2nd ed., McGraw-Hill, New York.
Ozgen, F.B. and Turan, A.H. (2007) Usage Adoption of Online Tax Filing Systems in Tax
Management: An Empirical Assessment with Technology Acceptance (TAM) Model in Turkey,
Accessed on 1st February, 2013, Http://Www.Getcited.Org/Pub/103442059
Ramlah, H., Norshidah, M., Abdul, R.A., Murni, M. and Umar, A. (2010) ‘An integrated model on
online tax adoption in Malaysia’, European, Mediterranean & Middle Eastern Conference on
Information Systems, (EMCIS2010), 12–13 April, Abu Dhabi, UAE, Available:
http://www.iseing.org/emcis/emcis2010
Sahu, G.P., Gupta, M.P. and Sahoo, T. (2004) ‘Towards a model of e-government acceptance’,
Proceedings of the 2nd International Conference on E-Governance, ICEG, 29 November–
1 December, Colombo, Sri Lanka, pp.1–8.
Saibal, S. (2007) E-Government Adoption Diffusion: National Informatics Centre, Bidyut Bhavan,
Saltlake, Kolkata, India,
Schwitzgebel, E. (2006) Belief’, in Zalta, E. (Ed.): The Stanford Encyclopedia of Philosophy,
The Metaphysics Research Lab, Stanford, CA, Retrieved 27th March, Http://Plato.
Stanford.Edu/Entries/Belief/
Senne, D. (2005) Ugandanda Moves toward E-Government, Accessed on 15th December, 2010,
www.Itweb.Co.Za/Sections/Enterprise/.Asp
Sserwanga, M. (2010) Ugandanda must Embrace the E-Government Model to Deliver Better
Services, Accessed on 20th February, 2011, Http://Blogspirit.Ug/Feed-Item
Tashakkori, A. and Teddlie, C. (2003) Handbook of Mixed Methods in Social Behavioral Research,
Sage, Thousand Oaks.
Uganda Radio Network (2010) Ura’s Online Tax Payment System Received with Mixed Reactions,
Http://Ugandaradionetwork.Com/A/Story
Uganda Revenue Authority (2012) E-Services, Accessed on January, 2011, Www.Ura.Co.Ug
United Nations E-Government Survey (2008) From E-Government To Connected Governance,
United Nations, New York, p.24.
Vassilakis, C., Lepouras, G., Fraser, J., Haston, S. and Georgiadis, P. (2005) ‘Barriers to electronic
service development’, E-Service Journal, p.41.
Venkatesh, V., Morris, M., Davis, G. and Davis, F. (2003) ‘User acceptance of information
technology: toward a unified view’, MIS Quarterly, Vol. 27, No. 3, pp.425–478.
Verlag, T. (2007) E-Taxation: State Perspectives, Series Informatics, Vol. 21, pp.195–206,
Accessed on 20th February, 2012, Http://Www.Ocg.At/Egov/Files/
Wang, Y.S. (2002) ‘The adoption of electronic tax filing systems: an empirical study’, Government
Information Quarterly, Vol. 20, pp.333–352.
Warkentin, M., Gefen, D., Pavlou, P. and Rose, G. (2002) ‘Encouraging citizen adoption
of e-government by building trust’, Electronic Markets, Vol. 12, No. 3, pp.157–162.
Woodgate, R. and Wilkins, K. (2008) ‘Designing a mixed methods study in pediatric oncology
nursing research’, Journal of Pediatric Oncology Nursing, Vol. 25, No. 1, 24–33.
Wougnet (2009) Call for External Evaluation on the ICT Usage Application by Women Farmers –
Northern Uganda, Accessed on 9th February, 2010, Http://Www.Wougnet.Org/
Yong, J.S.L. (2004) ‘Promoting citizen-centered approaches to e-government programs – strategies
& perspectives from Asian economies’, Second APEC High-Level Symposium on
E-Government, 6–8 October, Acapulco, Mexico, pp.356–367.
Yusniza, K. and Azmi, A.A.C. (2010) ‘Adoption of tax e-filing: a conceptual paper’, African
Journal of Business Management, Vol. 4, No. 5, pp.599–603.
Zarei, B., Ghapanchi, A. and Sattary, B. (2008) ‘Toward national e-government development
models for developing countries: a nine stage model’, The International Information Library
Review, Vol. 40, pp.199–207.

Maiga 2013

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Maiga 2013

Uploaded by

Copyright:

Available Formats

Electronic Government, An International Journal, Vol. 10, Nos.

3/4, 2013 259

Adoption of e-tax services in Uganda: a model of

Gilbert Maiga* and Elizabeth Asianzu

Abstract: Diverse benefits of e-government services are linked to its adoption

Keywords: e-government; e-tax adoption; e-government adoption models;

Reference to this paper should be made as follows: Maiga, G. and Asianzu, E.

Biographical notes: Gilbert Maiga has a PhD in Information Systems from

Elizabeth Asianzu holds a Bachelor of Administrative and Secretarial Science

Copyright © 2013 Inderscience Enterprises Ltd.

Whereas e-government adoption has increased for most countries, in economically

2.1 E-tax in Uganda

2.2 Challenges of e-tax adoption in Uganda

• inadequate IT human resource base

Figure 1 The technology acceptance model

Source: Davis (1989) and Venkatesh et al. (2003)

3.1 Modelling e-tax adoption with TAM: related work

Figure 2 A conceptual model for e-filing adoption in Malaysia

Source: From Yusniza and Azmi (2010)

Figure 3 The adoption of electronic tax filing systems model

Source: From Wang (2002)

Source: From Azmi and Bee (2010)

4 Deriving a model for e-tax service adoption in Uganda

4.1 The field study

4.2 Reliability of the questionnaire

Table 1 Summary of reliability of scales measures

Construct Cronbach’s alpha

4.3 Validity of the constructs

4.4 Field study results

4.5 A citizen-based e-tax adoption model

Figure 7 The citizen-based e-tax adoption model (extension of TAM)

Compatibility is a significant determinant in citizen’s intention to use e-government

Evaluation is the procedure of checking whether the model is applicable or not.

5.1 Reliability of validation questionnaire

Table 3 Reliability statistics of the validation questionnaire

Variable Cronbach’s alpha No. of items

variable that is accounted for by variation in the independent variables. It is measured by

Figure 8 Results of the regression analysis (significant relationships at p < 0.05)

5.3 Implications of the research

5.4 How to improve relatively low use of e-tax services

6 Conclusions and future work

A regression analysis provided support for relating the citizen-based factors

Ministry of Information Communications Technology (MoICT) (2009) Ministerial Policy

You might also like